PMLE Practice Test 20 — 15 Questions

Question 1

Which THREE are valid uses of Vertex AI Metadata?

Accepted Answer

Record the execution of a pipeline step. Vertex AI Metadata tracks lineage of artefacts, executions, and contexts across pipelines.

Answer

Deploy a model to an endpoint

Answer

Store hyperparameter values of an experiment run

Question 2

A team uses Vertex AI Pipelines and wants to track lineage of artifacts and executions. Which three resources should they use? (Choose three.)

Accepted Answer

Artifacts

Answer

Vertex AI Experiments

Answer

Model Registry

Question 3

A data science team uses Vertex AI Experiments to track training runs. They want to automatically log parameters, metrics, and artifacts for all runs with minimal code changes. Which approach should they take?

Accepted Answer

Use MLflow autologging by calling `mlflow.autolog()` before the training code and wrap the training script with `mlflow.start_run()`.. Vertex AI Experiments supports autologging via the MLflow library. By wrapping the training code with mlflow.start_run() and enabling autolog, all parameters, metrics, and artifacts are captured automatically.

Answer

Manually log each parameter and metric using `aiplatform.log_metrics()` after each training step.

Answer

Enable Vertex AI Experiments autologging by setting `autolog=True` in the experiment run context.

Answer

Use TensorBoard with tf.keras.callbacks.TensorBoard to log metrics.

Question 4

An ML engineer has a model trained in Vertex AI and wants to deploy it to an endpoint with autoscaling and traffic splitting for canary testing. They have the model artifact stored in Vertex AI Model Registry with alias 'champion'. What is the correct sequence of steps?

Accepted Answer

Upload model to registry, create endpoint, then deploy model to endpoint with traffic split.. The standard sequence: upload model to registry, create endpoint, deploy model to endpoint with traffic split.

Answer

Create endpoint, upload model to registry, then deploy model to endpoint with traffic split.

Answer

Create endpoint, deploy model directly from Cloud Storage, then add traffic split.

Answer

Upload model to registry, then create endpoint and deploy in one command using gcloud ai endpoints deploy-model.

Question 5

A company trains a model using features from Vertex AI Feature Store. They notice training-serving skew because the feature values used at training time differ from those served online. How should they address this?

Accepted Answer

Use point-in-time correct retrieval from the offline store for training data. Point-in-time correct retrieval ensures that the feature values used in training correspond to the exact same point in time as the label, preventing data leakage and skew.

Answer

Use the same online store for both training and serving

Answer

Disable caching in the online store

Answer

Enable feature monitoring to detect drift

Question 6

A data scientist is using Vertex AI Experiments to track training runs. They want to automatically log all hyperparameters, metrics, and model artifacts without modifying their training code. Which approach should they use?

Accepted Answer

Enable autologging with MLflow and use Vertex AI Experiments as the tracking server. Vertex AI Experiments supports autologging with MLflow integration. By using the MLflow tracking API or Vertex AI's autologging, parameters and metrics are captured without code changes.

Answer

Use Cloud Logging to capture training logs and parse them into experiments

Answer

Manually log parameters and metrics using the Vertex AI SDK in the training script

Answer

Run training in Vertex AI Workbench and manually save metrics to a BigQuery table

Question 7

A team uses Vertex AI Metadata to track pipeline runs. They need to identify all artifacts that were generated by a particular pipeline execution. Which API method should they use?

Accepted Answer

Use the lineage query API with the execution ID. The lineage query API can retrieve upstream and downstream artifacts for a given execution, allowing tracing of artifacts from a specific pipeline run.

Answer

List executions and then list artifacts separately

Answer

Create a context and query executions

Answer

Query artifacts by filter on execution ID

Question 8

An organisation uses Delta Lake on Dataproc to manage a data lake for ML training. They need ACID transactions for concurrent reads and writes. Which file format does Delta Lake use as the underlying storage?

Accepted Answer

Apache Parquet. Delta Lake uses Parquet as the base file format, adding a transaction log for ACID properties. Avro and ORC are not used by Delta Lake, and CSV does not support ACID.

Answer

Apache ORC

Answer

CSV

Answer

Apache Avro

Question 9

A team wants to enforce governance and compliance for all ML models across the organisation. They need a centralised repository that tracks model versions, deployment history, and evaluation metrics. Which service should they use?

Accepted Answer

Vertex AI Model Registry. Vertex AI Model Registry is the centralised repository for model versioning, lineage, metrics, and deployment management, providing governance and compliance.

Answer

Cloud Storage

Answer

Vertex AI Feature Store

Answer

Vertex AI Experiments

Question 10

You are performing post-training quantisation of a trained TensorFlow model to INT8 for deployment on edge devices. Which technique should you use to minimise accuracy loss?

Accepted Answer

Quantisation-aware training. Quantisation-aware training (QAT) simulates quantisation during training, allowing the model to adapt, resulting in higher accuracy than post-training quantisation. Post-training quantisation is simpler but may lose accuracy.

Answer

Float16 quantisation

Answer

Post-training integer quantisation with calibration

Answer

Post-training dynamic range quantisation

Question 11

A company wants to use Vertex AI Vizier to tune hyperparameters for a PyTorch model. They have a limited budget of 50 training jobs. The objective metric is validation accuracy, and they want to find the best configuration efficiently. Which algorithm should they choose?

Accepted Answer

Bayesian optimization using Vertex AI Vizier.. Bayesian optimization is the most efficient algorithm for hyperparameter tuning when the number of trials is limited. It builds a probabilistic model of the objective function and selects promising configurations.

Answer

Random search with 50 random configurations.

Answer

Use a custom algorithm implemented in the training code.

Answer

Grid search with 50 evenly spaced points.

Question 12

A data scientist wants to train a PyTorch model on Vertex AI using a pre-built container for GPU training. She needs to use 4 NVIDIA A100 GPUs on a single machine. Which machine configuration should she select?

Accepted Answer

a2-highgpu-4g (4 A100 GPUs). Vertex AI offers pre-built containers for PyTorch that support GPU training. To use 4 A100 GPUs, the machine type should be 'a2-highgpu-4g', which provides 4 A100 GPUs. The 'n1-standard-16' only supports up to 4 GPUs but typically uses P100 or T4, and 'n1-highmem-16' can support up to 4 GPUs but the GPU type is not A100 by default. The 'a2-megagpu-16g' provides 16 GPUs.

Answer

n1-highmem-16 with 4 NVIDIA V100 GPUs

Answer

n1-standard-16 with 4 NVIDIA T4 GPUs

Answer

a2-megagpu-16g (16 A100 GPUs)

Question 13

You have an edge device with limited compute resources. You need to deploy a deep learning model for real-time inference. Which model compression technique should you apply to reduce the model size and latency with minimal accuracy loss?

Accepted Answer

Post-training quantization to INT8. Post-training quantization (e.g., INT8) is the easiest and most effective method for reducing model size and latency on edge devices. Quantization-aware training can yield better accuracy but is more complex. Pruning and distillation also help, but quantization often gives the best trade-off.

Answer

Pruning only

Answer

Knowledge distillation only

Answer

Use full precision FP32 to maintain accuracy

Question 14

You are designing a distributed training job on Vertex AI for a PyTorch model using DataDistributedParallel (DDP). You have 4 nodes, each with 4 GPUs. What is the total number of workers that should be configured in the TF_CONFIG equivalent for PyTorch?

Accepted Answer

16. In PyTorch DDP on Vertex AI, each GPU typically runs one process. With 4 nodes each with 4 GPUs, the world size is 16 (total GPUs). Vertex AI distributed training uses TF_CONFIG to set up the cluster; for PyTorch, you need to set the number of workers to the total number of processes, which is 16.

Answer

4

Answer

8

Answer

1

Question 15

An organization is deploying a mission-critical model on Vertex AI Endpoints. They need to ensure high availability and meet a strict SLO of 99.9% uptime. Which THREE steps should they take? (Choose 3)

Accepted Answer

Set minReplicas to at least 2 to ensure redundancy within a region.. To meet a 99.9% SLO, they should deploy across multiple regions for redundancy, set minimum replicas to ensure baseline capacity, and configure health checks to route traffic away from unhealthy instances.

Answer

Use Cloud CDN to cache responses.

Answer

Use a single large instance instead of multiple small ones.